What problem does it solve?
Video teams struggle to manually index, search, and edit large libraries of footage across formats and live streams. This skill automates ingestion, indexing, and production workflows for video and audio content, enabling fast retrieval and playback. It coordinates server-side processing to extract frames, build visual/semantic/temporal indexes, and enable precise search with timestamps and auto-clips, while supporting real-time alerts from live streams or desktop capture.
Core Features & Use Cases
- Ingest media from local files, URLs, RTSP/live feeds, or live desktop captures.
- Build visual, semantic, and temporal indexes for fast search, clips, and evidence links.
- Transcode, normalize, and generate playable streams and alerts for live or recorded content.
- Create timeline edits (subtitles, overlays, branding, audio overlays, dubbing, translation) and generate media assets (image, audio, video).
- Use Case: Build a searchable video library from a conference recording and publish highlight streams.
Quick Start
Ingest a local video file, index spoken words and visuals, then generate a playable stream.